AMD Introduces Radeon R9 Fury Series Graphics Cards With Fiji GPUs

[dragonsqrrl]

Assuming the R9 nano is in the ~$300 price/performance range, that is the card I want. Too bad It won't come out until fall. I don't want to wait that long.

The Nano launches in the summer, the dual Fiji card is coming in the fall.

 

[dragonsqrrl]

If we are actually talking 4096 GCN cores with 1.5X the Performance per watt of Hawaii XT's 2816 GCN cores we are roughly talking about something with the power of a 295 X2 in a single GPU card that draws 275W. If this is all true we are seeing something that has never been seen before and that is a 100% performance increase from the previous generation.

I think your assessment is a bit off. I'm just not sure what brought you to the conclusion that a 45% increase in core count will result in a 2x performance increase. If it has a 275W TDP, which is probably a bit higher than the 290X, and it's a 50% improvement in perf per W with ~45% more compute resources, I'm guessing it's probably going to end up ~50% faster which should make it performance competitive with the 980Ti. This is all assuming that clock speeds haven't been compromised, but recent leaks have indicated that stock clocks will actually be higher than the 290X.

And a 2x performance increase over previous gen high-end is by no means unprecedented, it just hasn't happened recently.

 

[abhinav_mall]

400 A ???

 

[billgatez]

Anyone botherd by the fact that AMD did not talk about performance.

 

[rdc85]

The card looking sexy..

but I think their Fans for FuryX is overkill, if that fans is indeed Nidec D1225C (BB, B9, B7) it can pump up max 3000-5400rpm...
O. C. -er Dreams??? or maybe they will use same fans for the FuryX x2 card to make everything simpler...

 

[norseman4]

Assuming the R9 nano is in the ~$300 price/performance range, that is the card I want. Too bad It won't come out until fall. I don't want to wait that long.

The Nano launches in the summer, the dual Fiji card is coming in the fall.

While Tom's has been woefully wrong in the past, I haven't seen anything stating other than the following from the original article:

The Radeon water-cooled R9 Fury X will cost $649 and will be available on June 24. The air-cooled R9 Fury will sell for $549 and be on shelves a couple of weeks later on July 14. AMD did not reveal what Project Quantum and the Radeon R9 Nano would cost but said they will be available in the summer and fall, respectively.

Meaning Quantum will be in the summer and Nano will be in the fall. If you have conflicting sources, post them so Tom's could follow up and possible correct their article if they are wrong.

 

[photonboy]

4GB vs 6GB:
I see this as a huge problem for AMD. If their high-end cards only have 4GB and end up being comparable to the GTX980Ti 6GB.

As for all the hype over "HBM" stacked video memory etc... just remember it's a NON-ISSUE for most desktop users. The purpose of video memory is to have enough CAPACITY to store all the game data but be FAST enough so the GPU isn't waiting.

The new memory may have power savings but I'll take 6GB of GDDR5 over 4GB of the new stuff.

I'm not sure why people think the new video memory is the cats meow but its memory people.. not a new GPU architecture.

 

[norseman4]

Anyone botherd by the fact that AMD did not talk about performance.

Nope, this is the official announcement and they have made mention of some items but nothing reviewers, which can now get their hands on these cards, can start trying to break them, and report to both us, and AMD.

I'm looking forward to next week when we should start to see quality reviews, assuming that the usual suspects weren't bound by an NDA and legitimately didn't have versions of these cards already.

 

[somebodyspecial]

Yowza! We expected one new Fiji graphics card, but we got four new Fiji-based graphics products with HBM memory from AMD's latest announcement.

AMD Introduces Radeon R9 Fury Series Graphics Cards With Fiji GPUs : Read more

YOWZA...I was expecting actual product and benchmarks reviews everywhere...NOPE. What we got instead, is another delay for said info until what, like mid-end of july?...ROFL. This is not how you win a war AMD.

 

[somebodyspecial]

AMD has just hit a home run! With cards of such high grade performance at such low prices, it looks like they just gained the upper hand on Nvidia (I hope I'm not being too hasty in writing this. Benchmarks I guess will tell.)

ROFL...Where did that home run land? Oh right, there was no ball at all.

Way too hasty here...They have no hand currently. Your comment will be useful if they actually hit the ball and it leaves the park. Until those benchmarks you mentioned come, we have no idea if they even have a hit, let alone a home run.

 

[rdc85]

....said info until what, like mid-end of july?....

I read somewhere that Fury X is avail 24 June, Fury is 15 July, Nano and dual variant will be later ...
So we should see some benchmark soon..

Anyways, even there is some number from AMD marketing team..
it will be useless since it's AMD picked test, I prefer the one from tech site....

 

[jimmysmitty]

4GB vs 6GB:
I see this as a huge problem for AMD. If their high-end cards only have 4GB and end up being comparable to the GTX980Ti 6GB.

As for all the hype over "HBM" stacked video memory etc... just remember it's a NON-ISSUE for most desktop users. The purpose of video memory is to have enough CAPACITY to store all the game data but be FAST enough so the GPU isn't waiting.

The new memory may have power savings but I'll take 6GB of GDDR5 over 4GB of the new stuff.

I'm not sure why people think the new video memory is the cats meow but its memory people.. not a new GPU architecture.

The issue with GDDR5 is that it has hit its limits in bandwidth an right now games are pushing more textures so more bandwidth wont hurt.

Top it off with the fact that unless you plan to go into 4K with x8 MSAA you don't need more than 4GB of VRAM for most games. Hell even the Witcher 3 barely uses more than 2GB. That game just taxes systems.

HBM will be a major benefit. The next gen, which is coming next year, will double both capacity and bandwidth. So you will see 8GB GPUs with 1024GB/s of bandwidth which is a good thing as it will help alleviate one area so they can start to focus on other areas to improve. Both Pascal and a future version of AMDs Fury will be using HBM2 next year.

 

[dragonsqrrl]

Assuming the R9 nano is in the ~$300 price/performance range, that is the card I want. Too bad It won't come out until fall. I don't want to wait that long.

The Nano launches in the summer, the dual Fiji card is coming in the fall.

While Tom's has been woefully wrong in the past, I haven't seen anything stating other than the following from the original article:

The Radeon water-cooled R9 Fury X will cost $649 and will be available on June 24. The air-cooled R9 Fury will sell for $549 and be on shelves a couple of weeks later on July 14. AMD did not reveal what Project Quantum and the Radeon R9 Nano would cost but said they will be available in the summer and fall, respectively.

Meaning Quantum will be in the summer and Nano will be in the fall. If you have conflicting sources, post them so Tom's could follow up and possible correct their article if they are wrong.

Hmmm, unfortunately I haven't known Tom's to have the greatest track record when it comes to the accuracy of information in their news articles either. Actually that's being too generous, their news articles have had a terrible track record in my experience. Maybe that's changed recently, I wouldn't know. It also seems like the way the author worded the sentence could make it easy to get the two dates mixed up.

My source is from Anandtech's live blog from the show:
http://www.anandtech.com/show/9383/amd-radeon-live-blog
"01:04PM EDT - Nano in the summer, dual-GPU card in the fall"
"12:51PM EDT - Nano will be available later this summer"

I'm not claiming that Tom's is wrong, but if they're right, I would be surprised...

 

[fuzzion]

Benchmark shows the card is slower than the 980ti

http://videocardz.com/56225/amd-radeon-fury-x-3dmark-performance

 

[rdc85]

.... the card is slower than the 980ti....

Err what do you means "slower"?
in your linked "benchmark" FuryX is between titan X and above 980ti...

anyways that just "calculated" so not real bench...

 

[dragonsqrrl]

Benchmark shows the card is slower than the 980ti

http://videocardz.com/56225/amd-radeon-fury-x-3dmark-performance

The chart you linked shows it performing just about on par with the 980Ti/Titan X.

 

[dragonsqrrl]

.... the card is slower than the 980ti....

Err what do you means "slower"?
in your linked "benchmark" FuryX is between titan X and above 980ti...

anyways that just "calculated" so not real bench...

No, it's supposedly a real FireStrike result. He said he calculated the overall percentage as an average of the extreme and Ultra presets.

 

[norseman4]

My source is from Anandtech's live blog from the show:
http://www.anandtech.com/show/9383/amd-radeon-live-blog
"01:04PM EDT - Nano in the summer, dual-GPU card in the fall"
"12:51PM EDT - Nano will be available later this summer"

Yep, Anandtech probably has it right, which is cool.

 

[norseman4]

Benchmark shows the card is slower than the 980ti

If by slower, you actually mean faster, then the link that you posted agrees with you when you look at the text or the graphs.

 

[Wisecracker]

... The new memory may have power savings but I'll take 6GB of GDDR5 over 4GB of the new stuff.

I'm not sure why people think the new video memory is the cats meow but its memory people.. not a new GPU architecture.

Being that 4GB of HBM has 52% more bandwidth than the 12GB of GDDR5 on the TitanX, the HBM would have 2X the bandwidth of the 980ti.

Whoops.

When combined with the Tonga compression algorithms (never heard of those, huh?), the utilization and efficiency of that increased HBM bandwidth should increase something in the range of 35- to 40%.

Whoops. Again.

 

[TechyInAZ]

... The new memory may have power savings but I'll take 6GB of GDDR5 over 4GB of the new stuff.

I'm not sure why people think the new video memory is the cats meow but its memory people.. not a new GPU architecture.

Being that 4GB of HBM has 52% more bandwidth than the 12GB of GDDR5 on the TitanX, the HBM would have 2X the bandwidth of the 980ti.

Whoops.

When combined with the Tonga compression algorithms (never heard of those, huh?), the utilization and efficiency of that increased HBM bandwidth should increase something in the range of 35- to 40%.

Whoops. Again.

awww poor NVidia.

AMD defiantly has delivered well on it's GPU side. Nvidia will have to wait until next year for pascal to make up for what AMD is doing.

 

[oj88]

https://www.youtube.com/watch?v=1RdISWkIUmM demos the Quantum prototyping in AMD lab, very impressive.

 

[cobra5000]

Alright, alright, alright!

Matthew McConaughey

 

[clonazepam]

Firestrike scores will be artificially inflated due to the memory bandwidth that won't translate fully in games. Their claim of faster graphics card is likely true by comparing it against nvidia's cards at their reference clocks + unrealistic 3dmark performance. Once all of them are overclocked, then the argument of which is better just becomes academic.

I think the 390x / Titan X / 980 ti will rule over the fury x when doing multiple high resolution screens. Fury X will probably take the single up to 4K monitor crown in games with no hardware bias.

Does a win at 4K matter in 2015? Purely by the numbers and adoption rate, hell no. Next year is the year of 4K and VR. Pascal's already taped out, so hopefully nvidia doesn't have to spend too many millions on the masks at 16nm. That could be a bumpy and expensive road for them.

 

[somebodyspecial]

... The new memory may have power savings but I'll take 6GB of GDDR5 over 4GB of the new stuff.

I'm not sure why people think the new video memory is the cats meow but its memory people.. not a new GPU architecture.

Being that 4GB of HBM has 52% more bandwidth than the 12GB of GDDR5 on the TitanX, the HBM would have 2X the bandwidth of the 980ti.

Whoops.

When combined with the Tonga compression algorithms (never heard of those, huh?), the utilization and efficiency of that increased HBM bandwidth should increase something in the range of 35- to 40%.

Whoops. Again.

awww poor NVidia.

AMD defiantly has delivered well on it's GPU side. Nvidia will have to wait until next year for pascal to make up for what AMD is doing.

I think his point is there is no need for HBM when you have a gpu limited problem, not memory issue. You could have HBM rev69 out today, and it wouldn't speed AMD's cards up, on top of it you're using better compression FURTHER limiting the need for more bandwidth because (like maxwell) you're now using what you DID have far more efficiently.

See the point. NV has no need and thus avoided the costly memory. AMD probably should have done the same until the next gen. By using a new process AND a new memory type, they've likely killed all chance of profits and even yields for a bit (you have to get great yields on a new process and a new mem). I expect a bad quarterly report this Q, and next at least. Meanwhile NV will likely make a $150mil or so profit. Not sure you really understand what is going on here.